Session number: 1082
Abstract title: Changes in Length of Stay in VA and Private Sector Hospitals: Is VA Closing the Gap?
Author(s):
MJ Barnett - Iowa City VAMC
PJ Kaboli - Iowa City VAMC and University of Iowa
GE Rosenthal - Iowa City VAMC and University of Iowa
Objectives: Compare changes in recent years in length of stay (LOS) in VA and private sector hospitals for three groups: 1) VA patients in VA hospitals (VA); 2) VA patients receiving contracted care in private sector hospitals (VA contract); and 3) non-VA patients in private sector hospitals (private sector).
Methods: The sample included consecutive discharges over 4 years (1996-99) for 11 high–volume diagnoses (pneumonia, congestive heart failure, angina, coronary atherosclerosis, cardiac arrhythmia, chronic obstructive lung disease, gastrointestinal hemorrhage, urinary tract infection, diabetes mellitus, esophagitis, and metabolic disorders). Data for VA (n=440,441) and VA contract (n=12,857) patients were obtained from the Patient Treatment File. Private sector (n=19,066,807) data were obtained from the National Hospital Discharge Survey, a nationally representative sample of inpatients in non-Federal hospitals. Comorbidity was measured using a validated approach that considers 30 diagnoses that are unlikely to represent hospital complications.
Results: VA and VA contract patients were younger than private sector patients (mean ages 67, 66, and 70 years, respectively; p<.001) and more likely to be male (98%, 96%, and 42%; p<.001). VA patients tended to have higher comorbidity; compared to private sector patients, the prevalence was higher in VA patients for 8 of the 10 most common comorbid diagnoses. Unadjusted mean LOS in 1996 was higher for VA patients than private sector and VA contract patients (6.3 vs. 4.8 and 3.5 days, respectively). While unadjusted LOS remained higher in VA patients in 1997-99, differences decreased. In 1999 mean LOS in the 3 groups was 5.4, 4.5, and 3.7 days. These findings were also seen in multiple linear regression analyses adjusting for age, gender, and comorbidity: during the years 1996-1999, adjusted LOS in VA patients was 32%, 24%, 22%, and 19% higher, compared to private sector patients, and was 72%, 63%, 53%, and 39% higher, compared to VA contract patients.
Conclusions: While LOS has traditionally been longer in VA hospitals, our analyses suggest that the gap between VA and private sector LOS is narrowing.
Impact statement: These changes in LOS may reflect more aggressive utilization management practices in VA facilities that were implemented in the 1990’s and systematic efforts to shift care to outpatient venues.